minimally invasive esophagectomy: thoracoscopic mobilization of the esophagus and mediastinal...

MTMPCRR

CSPMTFCUCbcL

©P

inimally Invasive Esophagectomy:horacoscopic Mobilization of the Esophagus andediastinal Lymphadenectomy in Proneosition—Experience of 130 Patients

hinnusamy Palanivelu, MS, MCh, FRCS (Ed), FACS, Anand Prakash, MS, DNB, FNB (MAS),angaswamy Senthilkumar, MS, DNB, Palanisamy Senthilnathan, MS, DNB,amakrishnan Parthasarathi, MBBS, Pidigu Seshiyer Rajan, MS, FACS, S Venkatachlam, MD

BACKGROUND: To evaluate outcomes after minimally invasive or thoracolaparoscopic esophagectomy (TLE) withthoracoscopic mobilization of the esophagus and mediastinal esophagectomy in prone position.Esophagectomies are being performed increasingly by a minimally invasive route with decreasedmorbidity and shorter hospital stay compared with conventional esophagectomy. Most series reportthoracoscopic mobilization of the esophagus and mediastinal lymphadenectomy in the left lateralposition with respiratory complications up to 8% and prolonged operative time, probably because ofinadequate stance of the surgeon during the thoracoscopic part. This study shows the potential ofthe thoracoscopic part of the procedure in prone position to ease these difficulties.

STUDY DESIGN: From January 1997 through April 2005, TLE was performed in 130 patients. All patients hadhistologically proved squamous cell carcinoma of the middle third of the esophagus. Only one(0.77%) patient received neoadjuvant chemotherapy. The thoracoscopic part of the procedurewas performed in prone position with excellent ergonomics, translating into less operative timeand better respiratory results. We performed a minilaparotomy to retrieve the specimen owingto bulky tumors. Feeding jejunostomy and pyloromyotomy were performed in all patients.

RESULTS: There were 102 men and 28 women. Median age was 67.5 years (range 38 to 78 years). Therewas no conversion to open method. Median ICU stay was 1 day (range 1 to 32 days) and medianhospital stay was 8 days (range 4 to 68 days). Perioperative mortality was 1.54% (n � 2).Anastomotic leak rate was 2.31% (n � 3). There was no incidence of tracheal or lung injury anda very low incidence of postoperative pneumonia. At mean followup of 20 months (range 2 to70 months), stage-specific survival was similar to open and other minimally invasive series.

CONCLUSIONS: TLE with thoracoscopic part in prone position is technically feasible, with a low incidence ofrespiratory complications and less operative time required. It provides comparable outcomes withother techniques of minimally invasive esophagectomy and most open series. In our experience, weobserved a low mortality rate (1.54%), hospital stay of 8 days, and low incidence of postoperativepneumonia. It has the potential to replace conventional and other techniques of minimally invasive
esophagectomy. (J Am Coll Surg 2006;203:7–16. © 2006 by the American College of Surgeons)
Endoscopic Surgery, Hong Kong, August 2005; and 91st Annual Clinical Con-gress of American College of Surgeons, San Francisco, CA, October 2005.

Received October 5, 2005; Revised March 15, 2006; Accepted March 15,2006.From the Department of Minimal Access Surgery (Palanivelu, Prakash, Sent-hilkumar, Senthilnathan, Parthasarathi, Rajan, Venkatachlam) and SurgicalGastroenterology, GEM Digestive Disease Foundation (Palanivelu), GEMHospital, Coimbatore, India.Correspondence address: Chinnusamy Palanivelu, MS, MCh, FRCS (Ed), FACS,Department of Minimal Access Surgery, GEM Hospital, 45-A, Pankaja Mill Rd,

ompeting Interests Declared: None.upported by the GEM Digestive Disease Foundation, India.resented at the 8th World Congress of Endoscopic Surgery, New York, NY,arch 2002; Congress of Endoscopic and Laparoscopic Surgeons of Asia, 2002,

okyo, Japan, September 2002; Laparoscopic Approach to Cancer, Bordeaux,rance, October 2002; International Workshop on Foregut Surgery, Clevelandlinic, Miami, FL, February 2003; Advances in Complex Esophageal Surgery,niversity of Pittsburgh Medical Center, Pittsburgh, PA, July 2005; 9th Nationalonference of Laparoscopic and Endoscopic Surgery, Shanghai, China, Decem-er 2004; Indo-Japanese Conference of Minimally Invasive Surgery of GI Can-ers, Ahmedabad, India, July 2004; Demonstration of live procedure (Throraco-
aparoscopic Esophagectomy in Prone Position) at 7th Asia Pacific Congress of Ramanathapuram, Coimbatore 641045, India. email: [email protected]
72006 by the American College of Surgeons ISSN 1072-7515/06/$32.00

ublished by Elsevier Inc. doi:10.1016/j.jamcollsurg.2006.03.016

Ssdehpaftaitr

asancamsbnstrIFpia

neztsoaa

sp

MWvswaf

fwtrnwtdtft

OThsr

SeFealmplmtti

sstn

8 Palanivelu et al Minimally Invasive Esophagectomy J Am Coll Surg

urgical resection with curative intent is still the main-tay treatment of resectable esophageal cancer, one of theeadliest and least-studied cancers globally. Thesesophagectomies are performed by transthoracic, trans-iatal, or combined transthoracic and transhiatal ap-roaches. Traditional methods of esophagectomy aremong the most physiologically stressful procedures per-ormed in gastrointestinal tract surger y.1 These opera-ions are usually associated with considerable morbiditynd a mortality rate reported from experienced centersn the range of 6% to 7%.2 It has been a continuing questo search for the safest and least-traumatic method ofemoving the esophagus in managing esophageal cancer.

The blinding pace of progress in the field of minimalccess surgery, since the development of high-tech in-truments and equipment and the increasing expertise indvanced minimally invasive procedures, has openedew vistas to esophageal operations as well. These in-ludes laparoscopic fundoplication,3 laparoscopic man-gement of achalasia,4 giant paraesophageal hernia,5 andinimally invasive esophagectomies.6,7 Minimally inva-

ive esophagectomy has the potential to lower the mor-idity of open operations and allow quicker return toormal function.8,9 Several techniques have been de-cribed for minimally invasive approaches to esophagec-omy, eg, laparoscopic transhiatal,6 thoracoscopic,10 tho-acolaparoscopic,7 videomediastinoscopic,11 endoscopicvor Le wis,12 and laparoscopic esophagogastrectomy.13

or carcinoma of the middle third of the esophagus (es-ecially squamous cell carcinoma) combined approachs the most suitable to meet complete oncological clear-nce criteria.

Luketich and colleagues8 have pioneered the tech-ique of minimally invasive or thoracolaparoscopicsophagectomy (TLE), in which thoracoscopic mobili-ation of the esophagus is performed in left lateral posi-ion. We have developed a technique of thoracolaparo-copic esophagectomy with thoracoscopic mobilizationf the esophagus in prone position. We describe this newpproach, which has the advantage of better ergonomics

Abbreviations and Acronyms

ICS � intercostal spaceNG � nasogastricTLE � thoracolaparoscopic esophagectomy

nd has potential to replace conventional and thoraco- s

copic mobilization of the esophagus in left lateralosition.

ETHODSe store our data prospectively in an Institutional Re-

iew Board-approved database that was analyzed pro-pectively for outcomes after TLE. We used SPSS soft-are (version 10 for Windows; SPSS, Inc) to enter and

nalyze all data, and we used the Kaplan-Meier methodor survival analysis.

From January 1997 through April 2005, we per-ormed TLE in 130 patients, 102 men and 28 women,ith a median age of 67.5 years (range 38 to 76 years). In

his study, we included patients who were harboringesectable and histologically proved squamous cell carci-oma of the middle third of the esophagus. All patientsere evaluated and staged with preoperative CT scan of

he chest and abdomen and upper gastrointestinal en-oscopy. A complete evaluation of cardiac and respira-ory functions was made, and all patients were found fitor operation and anesthesia. Informed consent was ob-ained from all patients.

perative techniquehe operation is performed in three stages, as describedere. If preoperative assessment indicates doubtful re-ectability, we perform diagnostic thoracoscopy or lapa-oscopy, or both.

tage 1: Thoracoscopic mobilization of thesophagus and mediastinal lymphadenectomyirst, we perform thoracoscopic mobilization of thesophagus in prone position with right prone posteriorpproach. All patients were intubated with a single-umen endotracheal tube in the standard conventional

anner. After intubation, they were then turned torone position with beanbags under the chest and pelvis,

eaving the abdomen free to allow respiratory move-ents. All patients in this study underwent esophagec-

omy with single-lumen endotracheal tube with possiblewo-lung ventilation (left-lung ventilation with possiblentermittent ventilation of the right lung).

The right upper limb is abducted above 100°. Theurgeon stands on the right side of the patient, cameraurgeon to the left of the surgeon, and the second assis-ant stands on the left side of the patient. The scruburse stands on the left side of the patient. The laparo-
copic cart with monitor is positioned at the left shoul-

dc(trtTotptfrrisaas

MaAporllhaat

seiwttads

apcvsoecnrtLMnatt

LA

Fl

9Vol. 203, No. 1, July 2006 Palanivelu et al Minimally Invasive Esophagectomy

er of the patient. CO2 pneumothorax is created bylosed Veress technique in seventh intercostal spaceICS), later used as camera port, maintaining insuffla-ion pressure of 6 mmHg to 8 mmHg. The right lungemains collapsed throughout the operation because ofhis positive pressure and the additional help of gravity.hree ports are used: one 10 mm and two 5 mm. Oneptional 5-mm port can be used for additional retrac-ion. We use 30° scope for the whole procedure. Theor t positions (Fig. 1) are as follows: seventh ICS belowhe inferior angle of the scapula for camera (A): 10 mm,ifth ICS 7 cm right lateral from the spinous process foright-hand working (C): 5 or 10 mm, ninth ICS 7 cmight lateral from the spinous process for left-hand work-ng: 5 mm (B). A flexible upper gastrointestinal endo-cope can be placed into the esophagus to lift the esoph-gus to facilitate dissection. After general survey,dhesiolysis is performed, if present, by 5-mm harmonichears

obilization of the esophagus and control ofzygous veinzygous vein, esophagus with tumor-bearing area, andleura are inspected. Mobility of the tumor and invasionf the surrounding structure are then assessed. The infe-ior pulmonary ligament is divided to retract the inferiorobe of the lung. The mediastinal pleura above and be-ow the azygous vein are divided with electrocauteryook dissector and azygous vein is encircled completelynd doubly ligated and divided. The vascular stapler canlso be used to divide azygous vein. We prefer to ligate

Figure 1. Thoracoscopic port sites.

he vein by intracorporeal ligature. With the aid of an in i

itu endoscope, if necessary, the proximal esophagus isxposed and dissected circumferentially. Once a windown created in one position, the esophagus is encircledith umbilical tape. This tape is used for retraction of

he esophagus to facilitate dissection around it. By re-racting the esophagus ventrally, dissection is performednd the esophagus is dissected from the chest wall andescending thoracic aorta. Here, the thoracic duct isometimes visualized (Fig. 2) and has to be protected.

Dissection is continued between the esophagus later-lly and the superior vena cava medially.The mediastinalleura over the tumor and the entire esophagus are ex-ised en bloc. The vagal trunks are identified and di-ided. Then, dissection is commenced on the left sideupracarinally. The most difficult step is the separationf the membranous portion of the trachea from thesophagus. This is achieved by dissection with electro-autery hook and esophageal traction carefully coordi-ating the hand movement with deflation of the poste-ior wall of trachea at each expiration. Membranousrachea is at risk of injury at this point of dissection.aterally, dissection is carried down to the left pleura.obilization of the lower third of the esophagus poses

o problem. The periesophageal tissue and lymph nodesre included in the specimen. The entire esophagus fromhe thoracic inlet to the esophageal hiatus is mobilized athe end of this par t of the operation (Fig. 3).

ymph nodal clearancethorough nodal clearance is achieved initially in the

igure 2. Thoracic duct (A); top: thoracic aorta (B); bottom: mobi-ized esophagus (C) retracted by suction nozzle.

nfracarinal mediastinal space. The lymph nodes and

shhtortawantrantibrcgpitssrpwn

SsgTncauepl

F Fa

Fza


oft tissue are dissected from the caudal aspect near theiatus and proceeded cranially up to carina using 5-mmarmonic shears, avoiding injury to the pericardium an-eriorly and both bronchi on the sides. On completionf the lateral dissection, separation of remaining poste-ior attachments to the esophagus is completed. Here, ashe blood vessels are often small, dissection and controlre easily accomplished using an electrocautery hook,hich gives necessary traction and excellent maneuver-

bility. On the anterior side, once the inferior pulmo-ary vein is identified and dissected, a careful andhorough lymph nodal clearance is achieved in the sup-acarinal space. First, space between the right bronchusnd superior vena cava is dissected; soft tissue and lymphodes are swept and clearance is achieved up to thehoracic inlet. Next, trachea is pressed anteriorly us-ng left-hand instrument or by the assistant, and spaceetween left bronchus and arch of the aorta iseached; lymph nodes and soft tissue are dissected andlearance is performed along the left recurrent laryn-eal ner ve (Fig. 4) up to the thoracic inlet. This com-letes the dissection of the esophagus, including per-esophageal tissue and lymph nodes en bloc from thehoracic inlet to the esophageal hiatus (Fig. 5). Hemo-tasis is secured and an intercostal drainage tube is in-erted. The lung is allowed to expand fully, trocars areemoved, and ports are closed. At this stage, the patient’sosition is changed to modified Lloyd-Davis positionith reverse Trendelenburg for stage 2, ie, the abdomi-

igure 3. Thoracoscopic view of mobilized esophagus (A) at hiatus (B).

ocervical part of the procedure. i

tage 2: Laparoscopic mobilization of thetomach, removal of the specimen, andastric tube formationhe patient is positioned as described previously, witheck tilted toward the right side, exposing the left cervi-al area. Drapes are applied and whole abdomen, thorax,nd left cervical area are exposed. A total of five ports aresed: two ports of 5 mm each and three ports of 10 mmach. Positions of the por ts (Fig. 6) are as follows: su-raumbilical port for camera, 10 mm; epigastric port for

iver retraction, 5 mm; right midclavicular port for left-

igure 4. Left recurrent laryngeal nerve (A) seen after nodal clear-nce. Silk ligature can be seen on the cut end of the azygous vein.

igure 5. Thoracoscopic view after completing esophageal mobili-ation and nodal clearance. Center: membranous part of tracheand bronchi, pericardium; right: mobilized esophagus till thoracic
nlet.

hhg

tttoptmspegsapaTssopodsc

ic

fd(atpsaeaMtuiaaawscslcTNtcs

ScW


and working, 5 mm; left midclavicular port for right-and working, 5 or 10 mm; left anterior axillary port forastric retraction, 5 or 10 mm.

The surgeon performs the procedure standing be-ween the legs of the patient. The camera surgeon standso the right of the patient and assistant surgeon stands tohe left of the patient; scrub nurse stands to the rightf the surgeon. The laparoscopic cart with monitor islaced at the head end of the patient. The pneumoperi-oneum is created by closed Veress needle technique,aintaining 12 mmHg to 14 mmHg insufflation pres-

ure. The left lobe of the liver is retracted. The gastrohe-atic omentum is divided close to the liver and thesophagus is dissected from the right crus. Next, leftastric artery and vein are dissected and controlled withilk ligature or clips. The stomach is retracted anteriorlynd laterally. All lymphofatty tissue over the left gastricedicle, celiac axis, common hepatic artery, and splenicr ter y are dissected and vessels are skeletonized (Fig. 7).hen the gastrocolic omentum is divided. Proximal dis-

ection of the stomach is performed by dividing thehort gastric vessels using harmonic shears. The fundusf the stomach is dissected carefully from the superiorole of the spleen. The pancreaticogastric ligament andther attachments are divided up to the first part of theuodenum. This entails complete mobilization of thetomach. The pneumoperitoneum is deflated and cervi-al dissection is started.

Through a small 4-cm to 6-cm left skin crease cervicalncision, the cervical esophagus is dissected out in the

Figure 6. Abdominal port sites.

onventional manner. Umbilical tape is looped around it n

or traction and the esophagus is transected 2 to 3 cmistal to the upper esophageal sphincter. A nasogastricNG) tube is sutured to the distal cut end of the esoph-gus and the whole specimen is pushed into the medias-inum. The cervical wound is temporarily sealed withads. After creation of pneumoperitoneum, the entirepecimen is gradually pulled into the peritoneal cavitylong with NG tube under vision. The stomach and thesophagus along with NG tube are exteriorized throughminilaparotomy incision incorporating camera port.inilaparotomy incision of 4 to 5 cm is made just over

he antrum and a disposable plastic wound protector issed to prevent contact of specimen to the wound. This

ncision also facilitates pyloromyotomy. The stomachnd esophagus, along with NG tube, are placed in thenatomic position over a sterile sheet on the anteriorbdominal wall and chest. The specimen is transectedith formation of gastric tube using the linear-cutter

tapler or hand-sewn method, starting from the lesserurvature with minimum 5-cm margin. An adequatetretch is helpful to achieve satisfactory gastric tubeength. We prefer to make a wide gastric tube with 5m to 6 cm diameter added with a pyloromyotomy.he upper end of the gastric tube is stitched to theG tube (Fig. 8) and gastric conduit is returned to

he abdominal cavity. Minilaparotomy is closed andamera port is recreated. Now we proceed to the finaltage of the operation.

tage 3: Positioning of gastric conduit andervical anastomosise return to the abdominal cavity and pneumoperito-

Figure 7. Left gastric (A) and celiac axis (B) lymphadenectomy.

eum is reestablished. Under the laparoscopic guidance,

gattAsmTfceaptow

RO((hdppieptob

tlnttmrswp

(dTwwdsdclesdTdd(m

Fg


astric conduit is pulled through posterior mediastinumnd brought to the neck incision by careful traction onhe NG tube in the neck. Torsion is avoided by ensuringhat the greater curvature always lies toward the left crus.

wide esophagogastric anastomosis is performed in aingle layer in an end-to-end manner by the hand-sewnethod or using linear stapler in side-to-side manner.he tip of the gastric conduit is sutured to prevertebral

ascia. Neck wound is closed in layers after placing aorrugated drain. If redundancy is present at hiatus, neo-sophagus (gastric conduit) is sutured to diaphragm tovoid herniation (Fig. 9). A feeding jejunostomy tube islaced laparoscopically. One tube-drain is placed closeo hiatus, trocars are removed, and ports are closed. Allur patients were extubated on the table and ICU stayas required for intense monitoring only.

ESULTSf 130 patients, 102 (78.46%) were men and 28

21.54%) were women, with median age of 67.5 yearsrange 38 to 76 years). The preoperative indication wasistologically proved squamous cell carcinoma of mid-le third of the esophagus in all patients. One (0.77%)atient received neoadjuvant therapy. There was history ofrevious abdominal operations in 22 (16.92%) patients,

ncluding 6 (4.62%) with laparoscopic procedures. Preop-rative nutrition of these patients was maintained by high-rotein calorie liquid or soft diet as tolerated either orally orhrough endoscopically placed NG feeding tube. All TLEperations were performed at GEM Hospital, Coim-

igure 8. Positioning of gastric conduit after attaching it to naso-astric tube.

atore, India, by the same surgeon (CP). l

All procedures were completed by laparoscopy andhere were no conversions to open technique. Two-fieldymphadenectomy was performed in all patients. Meanumbers of lymph nodes harvested was 18 (range 11o 32). Mean operative time was 220 minutes (range 160o 450 minutes), and mean estimated blood loss was 180L (range 100 to 400 mL). Only 4 (3.08%) patients

equired peroperative blood transfusion. Median ICUtay was 1 day (range 1 to 32 days), time to oral intakeas 4 days (range 2 to 18 days), and postoperative hos-ital stay was 8 days (range 4 to 68 days).The 30-day postoperative mortality rate was 1.54%

n � 2). The first mortality was from an acute myocar-ial infarction occurring on the 9th postoperative day.he second death occurred in a 72-year-old male patientho showed signs of leak on the 4th postoperative day,hich was confirmed by Gastrografin (a mixture of so-ium amidotrizoate and meglumine amidotrizoate)wallow. Later, pleural effusion developed, which wasrained by changing the intercostal drainage tube. Heontinued to deteriorate, with appearance of right lowerobe pneumonia leading to ARDS on the 10th postop-rative day and was put on a ventilator. Subsequently, hehowed features of multiple organ dysfunction syn-rome and succumbed on the 18th postoperative day.here were no other deaths during the hospital stay (noeaths in patients who had stayed for more than 30ays). Postoperative morbidity encountered in 2720.76%) patients as outlined in Table 1. The most com-on major complications were anastomotic leak, de-

Figure 9. Fixation of the gastric conduit (A) to the diaphragm (B).

ayed gastric conduit emptying, and deep vein thrombo-

swptbmnc(ba1(m

DEc

tnshcawtpe

tpwtdfwnaulictilbaoesa

f

TM

AMDDCAPP

C


is (2.31% each). Pneumonia developed in two patients,hich worsened to ARDS in one patient. Postoperativeneumonia was defined as a positive bacterial culture ofhe sputum or abnormal chest radiograph with fever, oroth. Atrial fibrillation (5.38%) was the most commoninor complication. There was recurrent laryngeal

erve palsy in 2 patients (1.54%), which was treated byonservative methods. At mean followup of 20 monthsrange 2 to 70 months), 3-year survival of stages I, II a, II, and III were 75%, 50%, 45%, and 18%, respectively,s shown in the Kaplan-Meier sur vival cur ve in Figure0. In the postoperative quality-of-life assessment, 11689.23%) patients showed Visick score I, and the re-

aining 14 (10.77%) Visick score II.

ISCUSSIONsophageal resection in the management of esophagealancer has been a technical tour de force for conven-

able 1. Morbidity Data in Minimal Invasive Esophagectomyajor complications (n � 16 [12.30%]) n %

nastomotic leak 3 2.31yocardial infarction 2 1.54elayed gastric emptying 3 2.31eep venous thrombosis 3 2.31hylothorax 1 0.77cute cholecystitis 1 0.77neumonia 2 1.54leural effusion 1 0.77

omplications (n � 27[20.76%]).

nFigure 10. Kaplan-Meier survival curve.

ional and minimal access surgeons because of the tech-ical difficulties involved in the procedure and aggres-ive biologic behavior of the tumor, with consequentigh morbidity and mortality. As more approaches be-ome available for esophagectomy, the choice of typend access route continues to be a contentious issue ofhat is the optimal approach. The transthoracic and

ranshiatal approaches are the common surgical ap-roaches for management of the cancers of the thoracicsophagus.

Akiyama and colleagues14 have identified and mappedhe most common areas of lymph node metastasis fromrimary esophageal tumors. These nodal secondariesere then correlated with the location of the primary

umor to see whether certain primary tumors had ten-encies to spread to specific lymph node regions. Theyound out that one of the most common sites of spreadas the celiac axis. In middle third esophageal carci-oma, �50% of nodal metastasis occurred in posteriornd superior mediastinum and approximately 40% inpper abdominal nodes. The lymph nodal spread in the

ower third tumors was 60% in mediastinum and �70%n upper abdominal lymph nodes. Most reports, espe-ially from Japanese groups, focus on the value of ex-ended lymphadenectomy, both in the mediastinum andn the superior abdominal compartment (two-fieldymphadenectomy). Some surgeons think that addingilateral cervical lymphadenectomy (three-field lymph-denectomy)15 is also essential. With regard to the extentf the resection of esophagus, consensus is for subtotalsophagectomy because of the submucosal spread ofquamous cell cancers and frequent synchronous lesionnd multicentricity (�25%).16

We prefer middle and lower third esophageal cancersor surgical resection and chemoradiation for supracari-

Minor complications (n � 11 [8.46%]) n %

Atrial fibrillation 7 5.38Wound infection 2 1.54Recurrent laryngeal nerve injury 2 1.54

al and cervical lesions in our institution. For squamous

ctlrtlea

ttpsDtutsb3lhameamlt4ep3swdp

IriCttlppc

coOtrrwtlllpwptrefattvsipbctolilnfdd

esijwaaffd


ell carcinoma of the middle third esophagus, we selecthoracolaparoscopic esophagectomy enabling two-fieldymphadenectomy with subtotal esophagectomy. Lapa-oscopic transhiatal esophagectomy is preferred for lowerhird esophageal cancer of squamous type. Cuesta and col-eagues17 also endorse these vie ws. We opt for laparoscopicsophagogastrectomy and mediastinal esophagogastricnastomosis for adenocarcinoma of the cardia. 13

Since the first repor t by Cuschieri and colleagues,10

he minimally invasive or endoscopic approach has at-racted attention as a potentially less-invasive ap-roach to esophagectomy for esophageal cancer. Laparo-copic transhiatal esophagectomy was first described byePaula and colleagues6 from Brazil in a series of 12 pa-

ients. Their mean surgical procedure time was 256 min-tes (range 210 to 370 minutes), and mean postopera-ive hospital stay was 7.6 days (range 4 to 15 days). In aeries of laparoscopic total esophagectomy in 9 patientsy Swanstrom and colleagues,1 operation time averaged90 minutes (range 285 to 565 minutes), mean blood

oss was 290 mL (range 125 to 750 mL), and meanospital stay was 6.2 days (range 4 to 9 days). Nguyennd colleagues,18 in a series of 46 patients undergoinginimally invasive esophagectomy, reported mean op-

rative time of 350 minutes (range 210 to 520 minutes)nd mean blood loss of 279 mL (range 50 to 1,000 mL);edian length of ICU stay was 2 days, and median

ength of hospital stay was 8 days. Overall mortality inhis series was 4.3% and anastomotic leak rate was.35%. Luketich and colleagues8 have repor ted the larg-st series of minimally invasive esophagectomy in 222atients. In this series, anastomotic leak rate was 11.7%,0-day postoperative mortality was 1.4%, median ICUtay was l day (range 1 to 30 days), time to oral intakeas 4 days (range 1 to 40 days), and hospital stay was 7ays (range 3 to 75 days). Results of our study are com-arable with those of these series.

ncidence of pulmonary complications can beeduced by introduction of minimallynvasive approachuschieri and colleagues19 first repor ted a technique of

horacoscopic esophagectomy in prone position in 6 pa-ients and compared the results with those of leftateral position in 20 patients. They suggested that therone position has technical advantages and reducesostoperative respiratory complications. Luketich and
olleagues8 per formed the thoracoscopic par t of the pro- u
edure in left lateral position and reported the incidencef pneumonia in 7.6% and ARDS in 5% of patients.ur series reports lower incidence of pneumonia leading

o ARDS in 0.77% and pneumonia in 1.54%. Thiseduction in the pulmonary complications might be aesult of various factors. All patients in our study under-ent esophagectomy with single-lumen endotracheal

ube with possible two-lung ventilation (left-lung venti-ation with possible intermittent ventilation of the rightung). The partial or intermittent ventilation of the rightung reduces the venous shunt effect to a great extent. Inrone position, the functional residual capacity is morehen compared with functional residual capacity in su-ine position. Ventilation perfusion ratio is well main-ained and hypoxia and hypercarbia are avoided. Thiseduces the extent of pulmonary dysfunction and atel-ctasis postoperatively. Maintenance of ventilation per-usion ratio is evident from the fact that we did not haveny hypoxia and hypercarbia and we were able to main-ain oxygen saturation well within normal limitshroughout the operation. The possible intermittententilation of the right lung results in opening up aubstantial percentage of the alveoli, which might helpn prevention of postoperative atelectasis. This is notossible in the case of single-lung ventilation with a dou-le lumen endotracheal tube, as one lung is completelyollapsed throughout the entire procedure. Apart fromhe reduced pulmonary complications, there are alsother advantages to this approach. By using a single-umen endotracheal tube, we can avoid the problem ofntubation and maintain the position of the double-umen tubes, which at times is cumbersome. We also didot have any difficulty gaining access to the surgicalield; partial or intermittent ventilation of the right lungid not interfere with the surgical technique or prolonguration of the operation.The advantages of thoracoscopic mobilization of the

sophagus in prone position are mainly shorter anesthe-ia time; use of single-lumen endotracheal tube to allowntermittent inflation of right lung; decreased lung in-ury (the lung retraction is avoided), as the lung collapsesell because of the positive pressure pneumothorax,ided by gravity in prone position; decreased bronchialnd tracheal injuries; excellent exposure of the operativeield; and better ergonomics in stance of surgeon. Theseactors not only shorten total operative time but alsoecrease traumatic insult to the patient, and this might
ltimately lead to better postoperative respiratory func-

tcem

raeap1s6cslaTaa

iarpmspslpM1lcsast

ppnoaedpc

coSfofrnc

scrpwoasctertgpCi

TPa

NLW

*†


ions. The operative time in a previous Luketich andolleagues7 series of 77 cases of minimally invasivesophagectomy was 450 minutes, as compared with 220inutes in our series.Only a few studies have reported on outcomes of tho-

acolaparoscopic esophagectomy, as shown in Table 2. Inseries of 342 patients undergoing conventional 3-stagesophagectomy with right thoracotomy, laparotomy,nd cervical anastomosis, which is similar to our ap-roach, mortality of 3.6% and median hospital stay of3 days have been repor ted.20 They have repor ted 3-yearurvival for their stages I, II a, II b, and III patients as5%, 41%, 45%, and 17%, respectively. Luketich andolleagues8 have repor ted similar sur vival statistics. Oureries reports similar results, too. Nguyen and col-eagues9 have compared open transthoracic and transhi-tal approach to minimally invasive esophagectomy.his study favored the role of minimally invasive esoph-

gectomy with shorter operating time, less blood loss,nd shorter ICU and hospital stay in this group.

One disadvantage of a minimally invasive procedures reduced or lost ability of tactile identification, which isn important faculty to ensure an adequate margin ofesection. Use of intraoperative endoscopy solves thisroblem. None of our patients had positive transectedargins. There have been concerns about adequacy of

urgical margins, adequacy of lymphadenectomy, andort-site tumor implant in any laparoscopic or thoraco-copic procedure for cancer.21 We per formed two-fieldymphadenectomy, routinely including superior andosterior mediastinum and upper abdominal nodes.ean number of lymph nodes harvested was 18 (range

1 to 32) compared with 10.3 in Nguyen and col-eagues’18 series, 6 (range 3 to 12) in Swanstrom andolleagues’1 series, and 19 in Akaishi and colleagues’22

eries. We adhere strictly to all surgical principles tovoid port-site recurrence. We prefer removal of thepecimen through a minilaparotomy rather than

able 2. Outcomes for Selected Series of Thoracoscopic anublished series, firstuthor, year n

Mean operativetime (min)

Meanloss

guyen, 20009 18 364 29uketich, 20038 222 450* —atson, 200024 7 265† 20

From previous series of 77 cases.7

Median value.

hrough the narrow thoracic inlet as the majority of our r

atients harbor bulky tumor. We believe that this ap-roach has helped us prevent any recurrences in theeck. With a mean followup of 20 months, we have notbserved any wound or port-site recurrence. We havelways used wide gastric conduit of 5 cm to 6 cm diam-ter combined with pyloromyotomy, and gastric con-uit emptying and anastomotic leak have not been aroblem so far; this view is also favored by Luketich andolleagues.8

In quality-of-life assessment, we used Visick score be-ause of its simplicity and because of the unavailabilityf standard instruments such as the validated 36-Itemhort-Form Health Survey and gastroesophageal re-lux disease Health Related Quality of Life Scale inur local language (Tamil). We noted that at a meanollowup of 20 months, 89.23% showed Visick scoreemaining Visick and I score II. One advantage of aeck anastomosis is that incidence of reflux is lowompared with an intrathoracic anastomosis.8

We have demonstrated that TLE is a technically fea-ible procedure and can produce as good or better out-omes than most open series. We also believe that tho-acoscopic mobilization of the esophagus in proneosition is an ergonomically better technique comparedith that of left lateral position. It is important to pointut that our result of TLE is from a center with extensivedvanced and complex minimally invasive experience inurgical gastroenterology. TLE has the potential tohange the practice of transthoracic esophagectomy, al-hough this procedure is technically demanding. Thexperience of the surgeon is the determining factor toeduce morbidity in thoracoscopic esophagectomy.23 Ashe experience of a center increases in advanced esopha-eal operations, in conventional and minimally invasiverocedures, this might replace conventional operations.omparative and randomized, multiinstitutional stud-

es are needed to validate the procedure, to reproduce the

paroscopic EsophagectomyMean hospital

stay (d)Mortality

(%)Leak(%)

Conversion(%)

11.3 0 11 07* 1.4 11.7 7.2

12* 0 28.4 14.2

d Lablood(mL)

7

0*

esults, and to choose the best technique.

A

SAADCSO

S

R

1

1

1

1

1

1

1

1

1

1

2

2

2

2

2


uthor Contributions

tudy conception and design: Palanivelu, Prakashcquisition of data: Prakash, Seshiyer Rajannalysis and interpretation of data: Senthilkumarrafting of manuscript: Prakashritical revision: Palanivelu, Venkatachlamtatistical expertise: Senthilkumarbtaining funding: Palanivelu, Senthilnathan,Parthasarathi

upervision: Palanivelu

EFERENCES

1. Swanstrom LL, Hansen P. Laparoscopic total esophagectomy.Arch Surg 1997;132:943–947; discussion 947�949.

2. Kelsen DP, Ginsberg R, Pajak TF, et al. Chemotherapy followedby surgery compared with surgery alone for localized esophagealcancer. N Engl J Med 1998;339:1979–1984.

3. Dallemagne B, Weerts JM, Jehaes C, et al. Laparoscopic Nissenfundoplication: preliminary report. Surg Laparosc Endosc1991;1:138–143.

4. Luketich JD, Fernando HC, Christie NA, et al. Outcomes afterminimally invasive esophagomyotomy. Ann Thorac Surg 2001;72:1909–1912; discussion 1912�1913.

5. Pierre AF, Luketich JD, Fernando HC, et al. Results of laparo-scopic repair of giant paraesophageal hernias: 200 consecutivepatients. Ann Thorac Surg 2002;74:1909–1915; discussion1915�1916.

6. DePaula AL, Hashiba K, Ferreira EA, et al. Laparoscopic trans-hiatal esophagectomy with esophagogastroplasty. Surg LaparoscEndosc 1995;5:1–5.

7. Luketich JD, Schauer PR, Christie NA, et al. Minimally invasiveesophagectomy. AnnThorac Surg 2000;70:906–911; discussion911�912.

8. Luketich JD, Alvelo-Rivera M, Buenaventura PO, et al. Mini-mally invasive esophagectomy: outcomes in 222 patients. AnnSurg 2003;238:486–494; discussion 494�495.

9. Nguyen NT, Follette DM, Wolfe BM, et al. Comparison ofminimally invasive esophagectomy with transthoracic and trans-hiatal esophagectomy. Arch Surg 2000;135:920–925.

0. Cuschieri A, Shimi S, Banting S. Endoscopic oesophagectomythrough a right thoracoscopic approach. J R Coll Surg Edinb
1992;37:7–11.
1. Bonavina L, Incarbone R, Bona D, Peracchia A. Esophagectomyvia laparoscopy and transmediastinal endodissection. J Lapa-roendosc Adv Surg Tech A 2004;14:13–16.

2. Watson DI, Davies N, Jamieson GG. Totally endoscopic IvorLewis esophagectomy. Surg Endosc 1999;13:293–297.

3. Palanivelu C, Prakash A, Parthasarathi R, et al. Laparoscopicesophago-gastrectomy without thoracic or cervical access foradenocarcinoma of the gastro-esophageal junction: An Indianexperience from a tertiary center. Surg Endosc. In press.

4. Akiyama H, Tsurumaru M, Udagawa H, Kajiyama Y. Radicallymph node dissection for cancer of the thoracic esophagus. AnnSurg 1994;220:364–372; discussion 372�373.

5. Fujita H, KakegawaT, Yamana H, et al. Mortality and morbidityrates, postoperative course, quality of life, and prognosis afterextended radical lymphadenectomy for esophageal cancer.Comparison of three-field lymphadenectomy with two-fieldlymphadenectomy. Ann Surg 1995;222:654–662.

6. Kuwano H, Matsuda H, Matsuoka H, et al. Intra-epithelialcarcinoma concomitant with esophageal squamous cell carci-noma. Cancer 1987;59:783–787.

7. Cuesta MA, van den Broek WT, van der Peet DL, Meijer S.Minimally invasive esophageal resection. Semin Laparosc Surg2004;11:147–160.

8. Nguyen NT, Roberts P, Follette DM, et al. Thoracoscopic andlaparoscopic esophagectomy for benign and malignant disease:lessons learned from 46 consecutive procedures. J Am Coll Surg2003;197:902–913.

9. Cuschieri A. Thoracoscopic subtotal oesophagectomy. EndoscSurg Allied Technol 1994;2:21–25.

0. Swanson SJ, Batirel FF, Bueno R, et al. Transthoracic esophagec-tomy with radical mediastinal and abdominal lymph node dissec-tion and cervical esophagogastrostomy for esophageal carcinoma.Ann Thorac Surg 2001;72:1918–1924; discussion 1924�1925.

1. Johnstone PA, Rohde DC, Swartz SE, et al. Port site recurrencesafter laparoscopic and thoracoscopic procedures in malignancy.J Clin Oncol 1996;14:1950–1956.

2. Akaishi T, Kaneda I, Higuchi N, et al. Thoracoscopic en bloctotal esophagectomy with radical mediastinal lymphadenec-tomy. J Thorac Cardiovasc Surg 1996;112:1533–1540; discus-sion 1540�1541.

3. Osugi H, Takemura M, Higashino M, et al. Learning curve ofvideo-assisted thoracoscopic esophagectomy and extensivelymphadenectomy for squamous cell cancer of the thoracicesophagus and results. Surg Endosc 2003;17:515–519.

4. Watson DI, Jamieson GG, Devitt PG. Endoscopic cervico-thoraco-abdominal esophagectomy. J Am Coll Surg 2000;190:
372–378.

minimally invasive esophagectomy: thoracoscopic mobilization of the esophagus and mediastinal...

Documents